fix: update HuggingFaceEmbeddingEncoder to use `langchain_huggingface…

…` instead of `langchain-community` (#3436) Similar to #3433. ### Summary This PR aims to update `HuggingFaceEmbeddingEncoder` to use `HuggingFaceEmbeddings` from `langchain_huggingface` package instead of the deprecated version from `langchain-community`. This resolves the deprecation warning and ensures compatibility with future versions of langchain. ### Testing ``` from unstructured.documents.elements import Text from unstructured.embed.huggingface import HuggingFaceEmbeddingConfig, HuggingFaceEmbeddingEncoder embedding_encoder = HuggingFaceEmbeddingEncoder( config=HuggingFaceEmbeddingConfig() ) elements = embedding_encoder.embed_documents( elements=[Text("This is sentence 1"), Text("This is sentence 2")], ) query = "This is the query" query_embedding = embedding_encoder.embed_query(query=query) [print(e.embeddings, e) for e in elements] print(query_embedding, query) print(embedding_encoder.is_unit_vector(), embedding_encoder.num_of_dimensions()) ``` **Expected behavior** No deprecation warning should be displayed. The code should use the updated `HuggingFaceEmbeddings` class from the `langchain_huggingface` package.
Unstructured-IO · Jul 24, 2024 · 560cc0e · 560cc0e
1 parent 798dcc0
commit 560cc0e
Show file tree

Hide file tree

Showing 5 changed files with 21 additions and 86 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,11 +1,12 @@
-## 0.15.1-dev3
+## 0.15.1-dev4
 
 ### Enhancements
 
 ### Features
 
 ### Fixes
 
+* **Update `HuggingFaceEmbeddingEncoder` to use `HuggingFaceEmbeddings` from `langchain_huggingface` package instead of the deprecated version from `langchain-community`.** This resolves the deprecation warning and ensures compatibility with future versions of langchain.
 * **Update `OpenAIEmbeddingEncoder` to use `OpenAIEmbeddings` from `langchain-openai` package instead of the deprecated version from `langchain-community`.** This resolves the deprecation warning and ensures compatibility with future versions of langchain.
 * **Update import of Pinecone exception** Adds compatibility for pinecone-client>=5.0.0
 * **File-type detection catches non-existent file-path.** `detect_filetype()` no longer silently falls back to detecting a file-type based on the extension when no file exists at the path provided. Instead `FileNotFoundError` is raised. This provides consistent user notification of a mis-typed path rather than an unpredictable exception from a file-type specific partitioner when the file cannot be opened.

diff --git a/requirements/ingest/embed-huggingface.in b/requirements/ingest/embed-huggingface.in
@@ -1,5 +1,4 @@
 -c ../deps/constraints.txt
 -c ../base.txt
-huggingface
-langchain-community
-sentence_transformers
+
+langchain-huggingface
diff --git a/requirements/ingest/embed-huggingface.txt b/requirements/ingest/embed-huggingface.txt
@@ -4,20 +4,8 @@
 #
 #    pip-compile ./ingest/embed-huggingface.in
 #
-aiohttp==3.9.5
-    # via
-    #   langchain
-    #   langchain-community
-aiosignal==1.3.1
-    # via aiohttp
 annotated-types==0.7.0
     # via pydantic
-async-timeout==4.0.3
-    # via
-    #   aiohttp
-    #   langchain
-attrs==23.2.0
-    # via aiohttp
 certifi==2024.7.4
     # via
     #   -c ./ingest/../base.txt
@@ -27,36 +15,26 @@ charset-normalizer==3.3.2
     # via
     #   -c ./ingest/../base.txt
     #   requests
-dataclasses-json==0.6.7
-    # via
-    #   -c ./ingest/../base.txt
-    #   langchain-community
 filelock==3.15.4
     # via
     #   huggingface-hub
     #   torch
     #   transformers
-frozenlist==1.4.1
-    # via
-    #   aiohttp
-    #   aiosignal
 fsspec==2024.5.0
     # via
     #   -c ./ingest/../deps/constraints.txt
     #   huggingface-hub
     #   torch
-huggingface==0.0.1
-    # via -r ./ingest/embed-huggingface.in
 huggingface-hub==0.24.1
     # via
+    #   langchain-huggingface
     #   sentence-transformers
     #   tokenizers
     #   transformers
 idna==3.7
     # via
     #   -c ./ingest/../base.txt
     #   requests
-    #   yarl
 jinja2==3.1.4
     # via torch
 joblib==1.4.2
@@ -67,47 +45,21 @@ jsonpatch==1.33
     # via langchain-core
 jsonpointer==3.0.0
     # via jsonpatch
-langchain==0.2.11
-    # via langchain-community
-langchain-community==0.2.10
-    # via
-    #   -c ./ingest/../deps/constraints.txt
-    #   -r ./ingest/embed-huggingface.in
 langchain-core==0.2.23
-    # via
-    #   langchain
-    #   langchain-community
-    #   langchain-text-splitters
-langchain-text-splitters==0.2.2
-    # via langchain
+    # via langchain-huggingface
+langchain-huggingface==0.0.3
+    # via -r ./ingest/embed-huggingface.in
 langsmith==0.1.93
-    # via
-    #   langchain
-    #   langchain-community
-    #   langchain-core
+    # via langchain-core
 markupsafe==2.1.5
     # via jinja2
-marshmallow==3.21.3
-    # via
-    #   -c ./ingest/../base.txt
-    #   dataclasses-json
 mpmath==1.3.0
     # via sympy
-multidict==6.0.5
-    # via
-    #   aiohttp
-    #   yarl
-mypy-extensions==1.0.0
-    # via
-    #   -c ./ingest/../base.txt
-    #   typing-inspect
 networkx==3.2.1
     # via torch
 numpy==1.26.4
     # via
     #   -c ./ingest/../base.txt
-    #   langchain
-    #   langchain-community
     #   scikit-learn
     #   scipy
     #   sentence-transformers
@@ -120,22 +72,18 @@ packaging==23.2
     #   -c ./ingest/../deps/constraints.txt
     #   huggingface-hub
     #   langchain-core
-    #   marshmallow
     #   transformers
 pillow==10.4.0
     # via sentence-transformers
 pydantic==2.8.2
     # via
-    #   langchain
     #   langchain-core
     #   langsmith
 pydantic-core==2.20.1
     # via pydantic
 pyyaml==6.0.1
     # via
     #   huggingface-hub
-    #   langchain
-    #   langchain-community
     #   langchain-core
     #   transformers
 regex==2024.5.15
@@ -146,8 +94,6 @@ requests==2.32.3
     # via
     #   -c ./ingest/../base.txt
     #   huggingface-hub
-    #   langchain
-    #   langchain-community
     #   langsmith
     #   transformers
 safetensors==0.4.3
@@ -160,22 +106,17 @@ scipy==1.11.3
     #   scikit-learn
     #   sentence-transformers
 sentence-transformers==3.0.1
-    # via -r ./ingest/embed-huggingface.in
-sqlalchemy==2.0.31
-    # via
-    #   langchain
-    #   langchain-community
+    # via langchain-huggingface
 sympy==1.13.1
     # via torch
 tenacity==8.5.0
-    # via
-    #   langchain
-    #   langchain-community
-    #   langchain-core
+    # via langchain-core
 threadpoolctl==3.5.0
     # via scikit-learn
 tokenizers==0.19.1
-    # via transformers
+    # via
+    #   langchain-huggingface
+    #   transformers
 torch==2.3.1
     # via
     #   -c ./ingest/../deps/constraints.txt
@@ -187,24 +128,18 @@ tqdm==4.66.4
     #   sentence-transformers
     #   transformers
 transformers==4.43.1
-    # via sentence-transformers
+    # via
+    #   langchain-huggingface
+    #   sentence-transformers
 typing-extensions==4.12.2
     # via
     #   -c ./ingest/../base.txt
     #   huggingface-hub
     #   pydantic
     #   pydantic-core
-    #   sqlalchemy
     #   torch
-    #   typing-inspect
-typing-inspect==0.9.0
-    # via
-    #   -c ./ingest/../base.txt
-    #   dataclasses-json
 urllib3==1.26.19
     # via
     #   -c ./ingest/../base.txt
     #   -c ./ingest/../deps/constraints.txt
     #   requests
-yarl==1.9.4
-    # via aiohttp
diff --git a/unstructured/__version__.py b/unstructured/__version__.py
@@ -1 +1 @@
-__version__ = "0.15.1-dev3"  # pragma: no cover
+__version__ = "0.15.1-dev4"  # pragma: no cover
diff --git a/unstructured/embed/huggingface.py b/unstructured/embed/huggingface.py
@@ -11,7 +11,7 @@
 from unstructured.utils import requires_dependencies
 
 if TYPE_CHECKING:
-    from langchain_community.embeddings import HuggingFaceEmbeddings
+    from langchain_huggingface.embeddings import HuggingFaceEmbeddings
 
 
 @dataclass
@@ -69,12 +69,12 @@ def _add_embeddings_to_elements(self, elements, embeddings) -> List[Element]:
 
     @EmbeddingEncoderConnectionError.wrap
     @requires_dependencies(
-        ["langchain_community", "sentence_transformers"],
+        ["langchain_huggingface"],
         extras="embed-huggingface",
     )
     def create_client(self) -> "HuggingFaceEmbeddings":
         """Creates a langchain Huggingface python client to embed elements."""
-        from langchain_community.embeddings import HuggingFaceEmbeddings
+        from langchain_huggingface.embeddings import HuggingFaceEmbeddings
 
         client = HuggingFaceEmbeddings(**self.config.to_dict())
         return client
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		__version__ = "0.15.1-dev3" # pragma: no cover
		__version__ = "0.15.1-dev4" # pragma: no cover